The proposed research is designed to test a variety of theories that have been advanced to explain memory for the duration of an event in animals. The core phenomenon of interest is known as the "choose-short" effect. The choose-short effect refers to the fact that, no matter which duration was actually presented, pigeons tend to remember the duration as having been short. One well-known theory holds that the increasing tendency to report that the short sample occurred as the retention interval increases implies that the remembered duration of an event shrinks with the passage of time. Three other accounts of this phenomenon assume that the result stems from the fact that an event that is short in duration is more like "nothing" than an event that is long in duration. As such, when memories eventually fade to nothing (as they do when the retention interval is long), birds tend to select the better of the two available options (i.e., they report that the short sample was presented). Yet another explanation is that birds have the capacity to transform a nominal discrimination task into a signal detection task. The signal-detection framework has been used to interpret findings from studies on human recognition memory for more than two decades, but its relevance to animal memory has not been widely investigated. Because detection theory offers a theoretical framework common to both species, it could help to facilitate the development of relevant animal models of memory impairment in humans.